AVA
TOM TROSCIANKO AWARD
APPLICATION FORM
Applicants should be early career scientists (PhD students or within 5
years of obtaining their PhD) who are AVA members, or who are
sponsored by an AVA member. Funds will cover the cost of a visit to
ECVP and additionally must include a scientific visit to another location.
Applications will be judged on the scientific merit of the ECVP abstract,
and the additional visit.
_____________________________________________________________
1
Name (and title): Miss Cait Newport
Date of Birth: ***
Institutional Address:
(For all correspondence)
The University of Queensland
School of Biomedical Sciences
St Lucia, Queensland
Australia, 4072
E-mail address: c.newport@uq.edu.au
Telephone Number (day): ***
Present appointment: PhD candidate
_____________________________________________________________
2
Degrees and awards, with dates obtained. Those holding a PhD should
give the dates of submission and defence as well as the final award date. If
currently studying for a PhD please give details of topic and supervisor).
Bachelor of Science (hons) Marine Biology (Dalhousie University, Canada)
Completed: June 2007
Ph.D. Candidate (The University of Queensland, Australia)
Expected completion: January 2015
Supervisor: Dr Ulrike Siebeck
Project topic: Object recognition in fish
1
____________________________________________________________
3
The award covers travel to the European Conference on Visual
Perception and additionally MUST include a visit to at least one other
location at a time close to that of the ECVP meeting. The additional
activities can be for purposes of the candidate's choosing but they
should be of a scientific nature. Outline the details of your intended
visit(s), including (a) scientific purpose, (b) location(s) to be visited and
reasons why such visit(s) are essential to your scientific purpose. If
your plans include a laboratory visit or special access to other facilities,
please enclose a letter of agreement from the lab leader / facility
custodian:
Scientific purpose
As part of my PhD project, I am studying what visual information is required by
fish to discriminate complex patterns, such as the unique facial patterns found
on reef fish. In the past, it has been practically impossible to design visual
signalling experiments that comprehensively analyse patterns because natural
patterns are so complex. Attempting to test the individual importance of all the
visual information would require a vast number of experiments. Alternatively, a
limited number of salient features could be tested, however, selection of these
features can introduce significant experimenter bias. The advent of machine
learning, a branch of artificial intelligence, may provide a solution that will
eliminate the need to test every individual feature and instead identify
underlying mechanisms. Based on example behavioural data, this machine
learning can be used to mimic the behaviour of an animal, extract the learning
rules and use them to predict how the animal would respond to previously
untested stimuli, which can then be verified experimentally.
The application of machine learning to the study of animal sensory systems
has recently been pioneered by Prof. Franz to study the echolocation calls of
bats. I aim to apply this technique for the first time to the study of fish visual
perception and I will be visiting Prof. Franz at his laboratory in Konstanz,
Germany, to learn how it is done.
Locations to be visited
I will be travelling from Brisbane, Australia, to Zurich, Switzerland, where I will
then take a train to Konstanz, Germany, to visit The University of Applied
Sciences for three weeks. I will then fly to Hamburg and take a further train to
reach Bremen to attend the ECVP meeting.
Reasons for visit
The aim of my project is to look at how complex patterns are perceived by
fish. This project relies on an innovative combination of animal behaviour
experiments and modern engineering techniques and Prof. Franz is one of
very few people in the world who have the expertise for this ambitious project.
Prior to my visit I will have collected a large amount of behavioural data
describing the discrimination abilities of reef fish for conspecific facial patterns
that have been manipulated in various global ways (e.g. low and high-pass
2
frequency filtering, size transformations etc). In Konstanz, machines will be
trained to the same stimuli until their performance matches that of the fish.
The tricky part is then the extraction of the learning rules (reverse
engineering), which has only been done a few times before. Prof Franz is part
of the team who pioneered this technique and is therefore the best possible
person to approach to make this project work. By learning from Prof. Franz, I
will not only be able to apply his technique but also improve the design of
future experiments, as I will have a more fundamental understanding of what
can be achieved with machine learning approaches. The results of this work
will likely be applicable to a wide field, including the field of visual perception.
After my stay in Konstanz I will then travel to Bremen for the ECVP meeting
where I will have an opportunity to present results of a recent experiment
showing that archerfish can discriminate human faces. This will be a great
opportunity for me to learn about other research but it will also be a chance to
describe to others how fish can be used as a model to study visual perception
in an animal without a cortex.
_____________________________________________________________
4
Please provide a copy of your submitted ECVP abstract here:
There are currently two conflicting theories of how humans recognise faces: (i)
recognition processes are innate, relying on specialised cortical circuitry, and
(ii) recognition uses the same neural circuitry as other object classes and is
simply a learned expertise. One method to determine the underlying
mechanisms is to ask whether animals without specialised neural circuitry, or
indeed a cortex, can complete this task. We tested fish to determine whether
they could learn to discriminate human faces. Using a two-alternative forcedchoice test, four archerfish (Toxotes chatareus) were trained to select a
rewarded face image. All fish could select the correct face from 45 distractors
with an accuracy of over 75% (p<0.05). Humans tested using the same stimuli
reached a higher level of performance. However, archerfish performing a
much simpler task involving shapes (e.g. cross and square) revealed similar
levels of performance, suggesting that fish find human faces just as easy to
discriminate as shapes. This study provides the first behavioural evidence that
an animal lacking a cortex, and relatively little exposure to human faces, can
nonetheless discriminate them to a high degree of accuracy. Our results
suggest that a substantial part of the face discrimination task can be learnt.
5
Details of all proposed expenditure with fares, subsistence and other
expenditures clearly indicated
Item
Return airfare (Brisbane, Australia – Zurich, Switzerland)
Local transport (Konstanz – Hamburg, Germany)
Local transport (Zurich – Konstanz)
3
Total Cost
£2045
£200
£70
Local transport (Hamburg – Bremen)
Accommodation (Konstanz)
Accommodation (Bremen)
Subsistence
Total
£90
£336
£343
£1028
£2067
_____________________________________________________________
6
For what amount are you applying? (normally we expect to award up
to £1000 from AVA funds which may be topped up with additional funds
from ECVP, therefore requests exceeding £1000 will be considered
although we may offer partial funding):
I am applying for £1039 to cover the costs of accommodation and local
transport during my travels.
_____________________________________________________________
7
Other sources of assistance. Please give details of financial support
that is assured from other sources and applications that have been or
will be made.
My supervisor has paid for the return airfare from Australia to Switzerland and
will cover subsistence costs. Additionally, I have applied for a travel grant
(Graduate School International Travel Award) from The University of
Queensland but was unsuccessful.
_____________________________________________________________
I apply for a Tom Troscianko Travel Award from the Applied Vision
Association, and I declare that the details given above are correct to the best
of my knowledge.
I understand that a condition of the Award is to give a presentation about my
visit at the ECVP business meeting and to provide AVA with a report on the
ECVP conference attended, and the additional visit. The report will be sent to
the Secretary of the AVA, no later than 3 months after returning from the visit.
Signed
__***___________________
Date
_April 3, 2013____________________________
_____________________________________________________________
4
Closing date: 24th March 2013
We intend to inform applicants of the outcome no later than the early
registration deadline for ECVP
Email this file as an attachment to:
secretary@theava.net
Applied Vision Association Secretary
As well as this form, please ensure the AVA have been sent the following
documents before the closing date:
 If not an AVA member a sponsoring letter from an AVA member.
 If visiting another lab a letter of agreement from the host institution.
 If planning to access to a controlled facility a letter of from the
custodian agreeing to provide the required access.
5